# Homodyne radio

The invention relates to a homodyne radio receiving device. The technical result is the implementation of a homodyne radio receiver an improved estimation and compensation level DC (FRI-level). For this homodyne radio receiving device receives packets of radio frequency signals containing a number of symbols, and generates digital samples of at least a complex baseband signal (I and Q) of each packet signal. It includes a calculator averages, which determines the average PT is the level of the mentioned samples, and calculator sums of squares considering the sign of numbers, which calculates the sum of squares considering the sign of the numbers of many of the above samples and weighs the sum of squares considering the sign of the numbers using a compensation coefficient. Tool correction subtracts the sum of squares of the output signal calculator averages to estimate PT level. 6 N. and 13 C.p. f-crystals, 13 ill.

The scope to which the invention relates.

The present invention generally relates to a homodyne radio receiving device and in particular to a homodyne radio receiving device signals.

The level of technology

Homodyne receivers are well known and they are used in wireless communications devices, such as mobile phones.

Generally homodyne receiver contains two communication channels, which are usually known as I and Q channels. Dual-channel homodyne receiver is used in digital mobile phones, known from the prior art, such as GSM (Global system for mobile communications).

The receiver includes an antenna for receiving incoming electromagnetic communication signal, such as signal MDR (Multiple access with time division multiplexing), which is a digital stream of information symbols, which have been modulated in two orthogonal waves. The received signal is passed through band-pass filter, is amplified in the amplifier and then is divided into two identical signal. The first of these signals is in the first signal path, where he initially mixed in the mixer with the signal of the intermediate frequency. The intermediate frequency signal is supplied from the local oscillator and passes unchanged through the phase shifter. Similarly, the second signal is the second signal path, where it is mixed in the mixer with the signal of the intermediate frequency from heterocytes is filtered by the low pass filter and amplified in the second amplifier. Then the signal is in analog-to-digital Converter for sampling the signal and convert it into a digital signal that contains the stream of information symbols. The digital signal is filtered in a digital lowpass filter and digital information symbols contained in the signal into a digital memory. Set FR-level signal is determined and is subtracted from the digital signal, forming a digital signal, compensated with regard to FRI-level, which is demodulated.

The flow of digital information symbols are then used by other components in the mobile phone to obtain, for example, the sound output through the loudspeaker.

However, it was found that homodyne receivers, known from the prior art, are not high enough features, especially for informational messages sent between two computers during the session data.

Information frequency band homodyne receiver comes in frequency to PT, but PT-level does not contain information. Therefore, FRI-level must be removed before the information can be properly demodulated.

FRI-level signal can be evaluated according to what exploits calculator averages, which calculates the average value or the average PT-level digital signal in accordance with the following expression:

Therefore, the average value of the S election(i) is determined by obtaining the average value of a given number of N most recent samples taken. Then the average PT-level is subtracted from the digital signal, producing the digital signal is compensated with regard to FRI-level.

In systems mdvr similar to GSM, it is difficult to determine FRI-level, as the time for measuring PT-level is limited to one adopted by the packet signals, the received packet may have a different PT level because, for example, frequency-hopping. The number of samples contained in the calculation of the average values is, for example, 128 bits. Presumably this means that the average value of 128 GMM (Gaussian shift keying minimum shift modulated I samples (or Q samples) is zero. However, the average value of zero, is only the case when point signals are equally often in all quadrants in the IQ plane.

The study shows that the modulation causes, in some cases PT offset more than 20% (on the high ratio of N/N (carrier - the obstacle) and SNR (signal-to-noise) and an unacceptable number of errors in bits, especially when the signals are used to transfer data during a data session. Of course, the accuracy of estimation FRI-level also depends on the modulation type.

The invention

The present invention is the creation of a homodyne radio receiving device and method, which allows improved evaluation and compensation FR-level.

This task is solved by a homodyne radio receiver in accordance with the invention, according to which the package is accepted in the device radio frequency signals from a number of characters and is converted into digital samples for compensation through FRI compensator FRI. The first evaluation of the PT-level is determined by calculating the average PT-level samples in the calculator averages.

Then, in accordance with one aspect of the invention calculates the sum of squares considering the sign of the numbers of multiple samples through a calculator sums of squares considering the sign of the numbers in the compensator FRI. Sum of squaresweighed(using a compensation coefficient and the correction tool, coupled with a calculator, sums of squares, vicitim with another aspect of the invention, the tool compensation FR includes calculator amount and taking into account the sign of the numbers to calculate the sum with account of the sign of the numbers of many samples and weighting the sum by the coefficient of compensation. The remedy PT also provides a means of correction, and the output signal of the calculator amount considering the sign of the number is subtracted from the output signal of the calculator averages for FRI-level.

With the introduction of PT compensator, which assesses FRI-level, as described, and compensates the signal parameter depending on the shape of the signal, the output signal from the function will be correlated with the error in the estimation of PT. This function extracts information from the signal, which can be used to compensate for any error in the estimation FRI-level.

The advantage of the present invention is that estimation and compensation FR-level in accordance with the invention eliminates bounding the error in the PT layer with a high ratio of carrier/interference (n/a) and the signal-to-noise ratio (SNR) by incorporating information characteristic of the signal, when assessing FRI-level, leading to reduced the number of errors in the bits in the current conditions the signals.

Brief description of drawings

To explain in more detail the invention and its advantages and characteristics in the following detailed description of a preferred variant implementation, reference is made to attached the AI with the invention;

Fig.1B is a block diagram of a second variant implementation of the present invention in accordance with the invention;

Fig.2 is a graph of a probability function PT-error between -5 and +5%, depending on the angle in the equation of compensation;

Fig.3 is a graph of a probability function PT-error between -5 and +5%, depending on the angle in the equation of compensation;

Fig.4 is a graph of correlation for H/P>100 dB;

Fig.5 is a graph of correlation for N/N=6 dB;

Fig.6 is a graph of the probability functions for FRI error with compensation and without compensation for H/P>100 dB;

Fig.7 is a graph of a function compensated probabilities for different N/P;

Fig.8 is a graph of a function compensated probabilities for different N/P;

Fig.9 is a graph of the dependence of the correlation coefficient from within the counter.

Fig.10 is a plot of the error estimate PT from the difference between the number of negative and positive samples;

Fig.11 is a graph of the dependence of estimates with error -5 to +5% from the compensating angle;

Fig.12 is a graph of a probability function for FRI error with compensation and without compensation.

Detailed description of the invention

In Fig.1 shows a block diagram of one possible implementation of homodyne radio receiving device, in which ocenivaet demodulation signal.

The packet signal containing the number of characters that includes the carrier signal, the modulated baseband signal is received by the antenna 1 and is filtered by bandpass filter 2. The amplifier 3 amplifies the filtered signal generated by the filter 2. Then the enhanced packet signals are converted with decreasing frequency in the complex baseband signals In (in-phase) and Q (quadrature) conventional quadrature step-down Converter 4. The output signal I, Q down-Converter 4 is preferably filtered and amplified by additional filters and amplifiers before the output signal is sampled and converted into a digital signal by the analog-digital Converter (ADC) 5, 5. The digital signal may be filtered in a digital filter, and digital information symbols or samples contained in the signal, served in the digital memory 6 for later retrieval.

The output signal from the ADC 5, 5served in the device 7 compensation PT, which gives an estimate of the PT-level received signal and subtracts FRI certain level of packet signals before demodulation. To assess FRI-level compensation device PT is made with Voskanian. The output signal of the ADC 5, 5served in the calculator 8, 8average values, which evaluates the average PT-level series of samples. In addition, the calculator 8, 8averages detects the peak value of the signal by identifying the maximum value of the sample and subtracting the estimated average PT-level from the maximum value of the sample. The peak value is supplied to the next stage of the signal path together with an estimated average PT-level.

The next stage in the signal path is a calculator 9, 9sums of squares. Sampling the received packet signals stored in the digital memory 6 is read by the calculator 9, 9sum of squares, which standardizes the characters with reference to the estimated average PT level and the peak value before the sum of the squares considering the sign of the number of samples is calculated and weighted by the coefficient of compensation described below. Then calculates the evaluation of PT-level calculator 9, 9the sum of squared weighted subtraction of the calculated sums of squares considering the sign of the numbers.

Di and computing unit 10, 10that calculates a higher rating FRI-level by subtracting the calculated weighted sum of squares with account of the sign of the numbers of estimated average PT level. And finally, the increased valuation FRI-level is subtracted from each sample read from the memory 6. In an alternative embodiment of the invention, the subtraction of the calculated weighted sum of squares with account of the sign of the numbers of estimated average PT-level is carried out by the calculator 9, 9sums of squares.

Therefore, the output signal of the correcting and computing unit 10, 10at the end of the signal path will finally provide the first portion of the flow of digital information symbols contained in an analog signal, and transferred them, originally adopted by the antenna 1. The flow of digital information symbols are then used by other components in the mobile phone to obtain, for example, the sound output through the loudspeaker, for example speech. Or the flow of digital information symbols can represent informational messages sent between two computers during the session data.

The above limits degrade the performance of the device.

Therefore, in accordance with the invention, the error in the estimation FRI-level is minimized, if PT level is compensated by the information present in the packet signals.

In this embodiment of the invention FRI-level is assessed in accordance with the following expression (1) for compensation:

where

and

Thus the estimated PT-PT_{evaluation 1}is the average PT-level PT_{AVG.}; calculated for N symbols S(i) compensated on FRI error multiplied by the ratio of compensation. FRI error is estimated by calculating the sum of squares considering the sign of the numbers of the signal, in which x=S(i)-PT_{AVG.}

The optimum compensation coefficient, i.e. the ratio between PT-error and sum of squares considering the sign of the number is determined for different from what is shown in the graphs of Fig.2, the integrated area of a probability function PT-error between -5 and +5% is represented as a function of angle, i.e., the compensation coefficient in the expression (1). The solid line shows N/P>100 dB, dashed, N/N=20 dB, dashed line N/a=12 dB, and a double dotted line N/a=6 dB.

In Fig.3 shows graphs representing the integrated area of a probability function PT-error between -5 and +5%, as a function of angle, i.e., the compensation coefficient in the expression (1). The solid line depicts the SNR>100 dB, dashed, SNR=20 dB, dashed line SNR=12 dB and a double dotted line SNR=6 dB.

In accordance with the charts of Fig.2 and 3 it is obvious that the value of 2.0 is the optimal angle. Therefore, the ratio of compensation from 1.0 to -3,0 and mostly -2,0 is the optimal ratio between PT - error and sum of squares considering the sign of the numbers a high N/P. Thus the following expression is an example of an expression for compensation used by the device 7 compensation PT to PT evaluation-level.

The optimum linearity shows that this parameter is robust.

Fig.4 is a graph of correlation for H/P>100 dB is population will be more and more like a circle for the worst relationship N/a and therefore, then the optimal inclination will be, of course, less critical.

Probability functions for FRI error (as a percentage of the amplitude), estimated on the basis of the package of 128 characters for H/P>100 dB, shown in Fig.6. The solid line shows the probability function without compensation, and the dashed line is the probability function compensation amount of tilt is equal to 2 (compensation coefficient).

In Fig.7 shows the probability functions as the compensation in accordance with the invention for various N/a: solid line H/P>100 dB, dashed, N/N=20 dB, dashed line N/a=12 dB and a double dotted line N/a=6 dB.

In Fig.8 shows the function of the compensated probabilities for different SNR: solid line SNR>100 dB, dashed, SNR=20 dB, dashed line SNR=12 dB and a double dotted line SNR=6 dB.

Performance of the method in accordance with the invention is reduced when the signal interference (noise or other GMMS signal) is added to the desired signal. This is not a big problem, because performance will largely be determined by the fact that there is signal interference. At high SNR and N/a homodyne radio receiver in accordance with the invention, the error in the calculation of the PT level and the deterioration of the operation is progressing. This leads to a significant decrease in the number of bit-error even with a slight improvement in the calculation of the PT-level, which shows two graphs of Fig.6.

Therefore, by reviewing the information characteristics of the received packet signals and use them when evaluating FRI-level can be significantly reduced error in calculation FRI-level to high N/P ratio (carrier to interference) and SNR (signal/noise) signal.

Even if some signal interference or distortion from the filters of the receiver will degrade the functioning of the estimate PT-level in accordance with the invention will still be satisfactory for most applications.

The optimal slope is gentle, which gives the opportunity to evaluate FRI-level fixed value of the compensation coefficient, for example, -2,0.

It was shown that the difference between the number of samples with amplitude above/below a fixed limit of strongly correlated with PT-error and therefore it is used in the second embodiment of the invention.

As shown in Fig.1B, the packet signal containing the number of characters, and the carrier signal, the modulated baseband signal, p is qualified by the filter 12. Further enhanced packet signals are converted with decreasing frequency in the complex baseband signals I (inphase) and Q (quadrature) by conventional quadrature down-Converter 14. The output signals I and Q from the down-Converter 14 is preferably filtered and amplified extra filters and amplifiers before the output signal is discretized and converted into a digital signal by the analog-digital Converter (ADC) 15, 15. The digital signal may be filtered in a digital filter, and digital information symbols or samples contained in the signal, served in the digital memory 16 for later retrieval.

The output signal from the ADC 15, 15served in the device 17 compensation PT, which determines the score FRI-level received signal and subtracts FRI certain level of packet signals before it is demodulated. To determine the rating FRI-level device compensation FR processing of the received packet signals are made with the possibility of a large number of operations in accordance with the second embodiment of the invention. The output signal of the ADC 15, 15averages detects the peak value of the signal by identifying the maximum value of the sample and subtracting the estimated average PT-level from the maximum value of the sample. This peak value is supplied to the next stage of the signal path together with an estimated average PT-level.

The next stage in the signal path is the calculator 19, 19amount with account of the sign of the numbers. Digital information symbols or samples of the received packet signals stored in the digital memory 16 is read by the calculator 19, 19amount due to character numbers, which standardizes these symbols with reference to the estimated average PT level and the peak value before the sum considering the sign of the number of samples is calculated and weighted by the coefficient compensation, as described below.

The output signal of the calculator 19, 19amount with account of the sign of the numbers served in correcting and calculating block 20, 20that calculates improved renego FRI-level. And finally, improved estimation FRI-level is subtracted from each sample read from the memory 16.

Then the output signal of the correcting and calculating unit 20, 20at the end of the signal path will provide finally, the flow of digital information symbols contained in an analog signal and transported analog signal, originally adopted by the antenna 1, but compensated with regard to FRI-level, obtained from the estimation, as described above.

Thus, in this embodiment of the invention FRI-level is assessed in accordance with the following expression (5) for compensation:

where

and

Thus the estimated PT-PT_{assessment 2}is the average PT-level PT_{AVG}., calculated on N symbols S(i), compensated with regard to FRI-errors FRI_{error 2}that is multiplied by a factor In compensation. FRI error is estimated by calculating the sum signal, taking into account the sign of numbers, calculated by M symbols S(j), where y=S(j)-PT_{AVG.}

A plot of the correlation coefficient from within schetchik-50% and preferably 40% amplitude are used to calculate the difference between the number of positive and the number of negative samples. The error in the evaluation of the PT layer is that the sample close to the PT-level, unreliable, and therefore preferred a higher limit. However, a higher limit reduces the number of samples used for calculation. As mentioned above, the flatness of the optimum shows that the compensation coefficient is robust.

The optimal rate of reimbursement is based on different relations N/a (carrier/interference) and SNR (signal/noise), suggesting the presence of white noise.

In Fig.10 shows the error in the estimate of PT as a percentage of the amplitude as a function of the difference between the number of negative and positive samples, and Fig.11 shows the estimated error in the range of -5% to +5% as a function of the compensation angle.

From the graph in Fig.10 obviously, the compensation coefficient value 0,0055-0,0065 and mostly 0,006 is the optimal ratio between PT-error and amount considering the sign of the numbers. Thus the following expression is an example of compensatory expression used by the device 4 compensation PT assessment FRI-level, if the computation uses only the samples (in the expression (6)) above/below the 40% range.

Probability functions for FRI error (as a percentage of the amplitude), estimated with the use of a package of 128 characters for N/a>1001 dB shown in Fig.12. The solid line shows the probability function, without compensation, the dashed line shows the probability function with the value of the compensation angle 0,006.

Therefore, by accounting information characteristics of the received packet signals and the use of samples above/below a certain percentage of the amplitude estimates for FRI-level error in the determination FRI-level can be significantly reduced for high relations N/a and SNR conditions the signal to pass through.

Although the invention has been described through specific examples of its implementation, it should be apparent that the present invention provides an improved method of estimating FRI-level and homodyne radio receiving device which fully meets the objectives and advantages set forth above, and their alternatives, modifications and variations obvious to a person skilled in this technical field.

The compensation device PT is a digital device for signal processing, preferably implemented in the form of diagrams with rigid logic for fast functioning. But is exploring the embodiment of the invention.

Other systems, such EDGE (electronic data acquisition system) and AMR (automatic message recording), have some coding scheme, which provides a good condition for a signal. One single-bit error causes to repeat the transfer of a large block, and therefore it reduces the bandwidth when transferring data.

Claims

1. Homodyne radio receiving device containing a homodyne radio receiver(1, 2, 3, 4, 5, 5for packet reception of radio frequency signals from a number of characters and forming digital samples of at least complex baseband I and Q of each of the packet signals and the means (7) compensation direct current (DC), United with a homodyne radio receiver(1, 2, 3, 4, 5, 5) assessment FRI-level above-mentioned samples, and means (7) compensation includes the computing device (8, 8average values to calculate the average PT-level above-mentioned samples, characterized in that the means (7) compensation FR further comprises a computing device (9, 9) sum of squares taking into account the mentioned sum of squares considering the sign of the numbers using a compensation coefficient, and means (10, 10) correction for subtracting the output signal of the computing device sums of squares (9, 9) from the output signal of the computing device (8, 8average values to obtain these ratings FRI-level.

2. Homodyne radio receiver under item 1, characterized in that the said compensation coefficient ranges from -1 to -3, and preferably -2.

3. Homodyne radio receiving device containing a homodyne radio receiver(11, 12, 13, 14, 15, 15for packet reception of radio frequency signals containing the number of characters, and forming digital samples of at least a complex baseband signal (I and Q) of each packet signal, and means (17) compensation direct current (DC), United with a homodyne radio receiver(11, 12, 13, 14, 15, 15) assessment FRI-level above-mentioned samples, and means (17) compensation FR contains a computing device (18, 18average values to calculate the average PT-level above-mentioned samples, characterized in that the means (17) compensation THU dopolnitelnaya amount considering the sign of the numbers of many of the above samples and weighing the stated amounts by using a compensation coefficient, and means (20, 20) correction for subtracting the output signal of the computing device (19, 19) amount considering the sign of the numbers from the output signal of the computing device (18, 18average values to obtain these ratings FRI-level.

4. Homodyne radio receiving device according to p. 3, characterized in that the said samples have a value of amplitude of at least 40% of the mentioned amplitude of the modulating signal.

5. Homodyne radio receiving device according to p. 4, characterized in that the compensation coefficient is 0,0055-0,0065 and preferably 0,006.

6. Homodyne radio receiving device according to any one of paragraphs.1-5, characterized in that the said means (7; 17) compensation FR is a device for digital signal processing.

7. Homodyne radio receiving device according to any one of paragraphs.1-5, characterized in that the homodyne radio receiver includes means(1, 2, 3; 11, 12, 13) for receiving signals, connected to the quadrature step-down Converter (4; 14) for conversion downconverter mentioned package signals to at least a complex baseband signal (I and Q), and analog-to-digital Converter is.

8. Homodyne radio receiving device according to any one of paragraphs.1-5, characterized in that the said means (10, 10, 20, 20) correction made with the possibility of subtracting the above-mentioned PT evaluation-level of each sample.

9. A signal processor to compensate for the level of direct current (FRI-level packages of radio frequency signals received by a homodyne radio receiver containing means for receiving signals for receiving digital samples of at least a complex baseband signal (I and Q), mentioned formed a homodyne radio receiver means (7) compensation assessment FRI-level above-mentioned samples, the computing device (8, 8average values to calculate the average PT-level above-mentioned samples, characterized in that it contains a computer device (9, 9) sum of squares considering the sign of the numbers to calculate the sum of squares considering the sign of the numbers of many of the above samples and weighing mentioned sums of squares using the compensation coefficient, and means (10, 10) correction for subtracting the output signal referred to the high unit (8, 8average values to obtain these ratings FRI-level.

10. The signal processor under item 9, characterized in that the said compensation coefficient ranges from -1 to -3, and preferably -2.

11. A signal processor to compensate for the level of direct current (FRI-level packages of radio frequency signals received by a homodyne radio receiver containing means for receiving signals for receiving digital samples of at least a complex baseband signal (I and Q), mentioned formed a homodyne radio receiver, means (17) compensation PT assessment FRI-level above-mentioned samples, the computing device (18, 18average values to calculate the average PT-level above-mentioned samples, characterized in that it contains computing device (19, 19) amount considering the sign of the numbers to calculate the sum with account of the sign of the numbers of many of the above samples and weighing the stated amounts by using a compensation coefficient, and means (20, 20) correction for subtracting the output signal of the aforementioned computing devices (19, 19) su the border="0">average values to obtain these ratings FRI-level.

12. The signal processor according to p. 11, characterized in that the said samples have a value of amplitude of at least 40% of the mentioned amplitude of the modulating signal.

13. The signal processor under item 12, characterized in that the said compensation coefficient is 0,0055-0,0065 and preferably 0,006.

14. The signal processor according to any one of paragraphs.9-13, characterized in that the said homodyne receiver includes means(1, 2, 3; 11, 12, 13) for receiving signals, connected to the quadrature step-down Converter (4; 14) for conversion with decreasing frequency of the above-mentioned packet signals at least in the complex baseband signal (I and Q), and analog-to-digital Converter (5, 5; 15, 15for forming the above-mentioned samples.

15. The signal processor according to any one of paragraphs. 9-13, characterized in that the said means (10,10; 20, 20) correction made with the possibility of subtracting the above-mentioned PT evaluation-level of each sample.

16. The method of assessing the level of direct current (FRI-level) in a homodyne radio receiving device, irout digital sampling at least a complex baseband signal (I and Q) of each packet signals, determine an estimate of the average PT-level above-mentioned samples, characterized in that it further determine the sum of squares considering the sign of the numbers of many of the above samples and weighed mentioned the sum of squares using the compensation coefficient and subtract the sum of the squares considering the sign of the numbers of the above-mentioned medium-FRI-level to estimate referred to FRI-level.

17. The method according to p. 16, characterized in that the said compensation coefficient ranges from -1 to -3, and preferably -2.

18. The method of assessing the level of direct current (FRI-level) in a homodyne radio receiving device, namely, that take in the device package RF signals containing the number of characters that form a digital sample at least a complex baseband signal (I and Q) of each packet signals and determine an average rating FRI-level above-mentioned samples, characterized in that which further determine the amount considering the sign of the numbers of many of the above samples and weigh the amount considering the sign of the numbers using a compensation coefficient and subtract the amount from the account of the sign of the numbers of the above-mentioned medium-FRI-level to obtain these ratings FRI-level.

**Same patents:**

FIELD: controlling flying vehicles by optical beam from control stations in optical communications and television systems.

SUBSTANCE: proposed method for decoding pulse-width modulated and frequency keyed harmonic-wave signal (guidance command) on board missile used to convert electromagnetic radiation into harmonic signal includes generation of rectangular pulses from harmonic signal filtered off at operating frequency whose amplitude exceeds preset passage threshold, their length being determined by double cutoff angle and separation of constant component from rectangular pulses; in addition harmonic signal converted from electromagnetic radiation is symmetrically limited relative to zero, amplified, and converted into rectangular pulses of duty factor two, then harmonic signal is filtered off from them at operating frequency.

EFFECT: enhanced precision of decoding due to eliminating impact of modulation noise.

2 cl, 1 dwg